The Trust for Thanet Archaeology: Evaluation Work carried out in 1995, Hartsdown Community Woodland Scheme, Margate
THE TRUST FOR THANET ARCHAEOLOGY;
EVALUATION WORK CARRIED OUT IN 1995,
HARTSDOWN COMMUNITY WOODLAND SCHEME,
MARGATE
D. R. J. PERKINS
INTRODUCTION
This archaeological evaluation survey on farmland at Hartsdown/Shottendane,
Margate, was made necessary by plans to develop the land as a
Community Woodland. Extensive tree-planting is intended, and since a
number of crop-marks representing ancient activity and settlement
appear on the fields, their evaluation was essential. The evaluation survey
demonstrated that a number of archaeological sites visible as crop-marks
in aerial photographs taken some thirty years ago were present in a good
state of preservation, and were of such interest as to form an important
factor in planning a community woodland project for the Hartsdown site.
The concept of a project to establish a Community Woodland on
Thanet Council owned land at Hartsdown/Shottendane, Margate, was
first mooted in 1991. The Planning Department of Thanet District Council
then asked the Thanet Archaeological Trust for advice as a matter
of routine, and the Trust was able to state that there were important
archaeological implications attendant on the proposed development.
Aerial photographs taken in the 1960s demonstrated the presence of a
number of enclosures that could be interpreted as the ring ditches of
prehistoric round bru.Tows, and prehistoric and/or Romano-British settlements.
The Trust began a study of the crop-marks using the 1960s photos with
some more recent colour slides in conjunction with a computer-aided
photogrammetry programme to plot the crop-marks on Ordnance Survey
maps. An evaluation strategy consisting of a pattern of random and 'educated'
trenches was designed, and this was submitted for approval to the
County Archaeologist, Dr John Williams. The evaluation was carried out
in August 1995, and record, photographic, and material archives of the
work have been created.
265
D. R. J. PERKINS
ACKNOWLEDGEMENTS
The Trust for Thanet Archaeology wish to thank Thanet District Council
for funding this important evaluation, and Mr Jonathan Tapp who farms
the land, both for his kind co-operation, and careful attitude to cultivating
the land so that the ancient sites have been largely preserved from plough
damage.
THE SITE TOPOGRAPHY AND KNOWN ARCHAEOLOGICAL AND HISTORICAL
BACKGROUND
Site Topography and Geology
The site is situated at Hartsdown/Shottendane, Margate, on farmland part
of Twenties Farm at an O.S. ref. centred at TR 345 695. It is 20.62 ha.
(50.9 acres) in area. The fields constituting the site form the northern
rise and escarpment ridge of the Shottendane Valley, and vary in elevation
between 10.50 and 25.00 m. O.D. There can be little doubt that
the valley held a stream until modern times, its last remnant being the
'Tivoli Dike' which last century drained into the marsh where Margate's
'Dreamland' Amusement Park is now situated, and from which Margate
(Marshgate) is said to have been named. In 1808, during a storm and
on-shore gale, the valley for a few hours became a tidal creek as far up
as Shottendane Farm.
The geology of the site is that of the Head and Drift Brickearth over
the Upper Chalk. Since this stratum drains well the Shottendane stream
may only have run in winter, reverting to a series of ponds and marshy
spots in high summer. In this it would have been similar to a number of
Kent's 'woe waters' as for example the upper reaches of the Nailbourne.
Three of the evaluation trenches, Trenches 37, 38, and 47, were cut
near the valley bottom where they encountered colluvium so deep that
they could not safely be excavated down to chalk. Elsewhere on the site
trenching revealed the chalk to be capped by an overburden of loam
varying between 0.30 and 0.60 m. in depth, of which the upper 0.25 m.
is modem plough soil with that beneath it constituting an ancient horizon
often exhibiting cultural material. The level of the Roman metalled trackway
surface (Site 1) and ditch section evidence from Site 7 seem to
indicate that ground level today is little different from what it was in the
Iron Age and Roman periods. Over much of the site the surface of the
Upper Chalk was hard and even. In the northern angle near Hartsdown
Road it was, however, much affected by periglacial activity, the brickearth
filled ice-wedge pits being easily confused with man-made features.
That ancient horizons survive at Hartsdown when they have been
266
HARTSDOWN COMMUNITY WOODLAND SCHEME, MARGATE
denuded elsewhere in Thanet is due to the care of the farmer, Mr Jonathan
Tapp, who after attending an N.F.U. branch meeting some years
ago at which the writer demonstrated the consequences of deep ploughing
over archaeological sites, has eschewed the use of the 'sub-soiler'.
Known Archaeology
No archaeological investigations even at the level of field-walking have
previously taken place at Hartsdown. There has been a certain amount
of metal detector prospecting around the margins of the field, albeit with
little of interest found (pers. comm. Thanet and Wantsum Association).
The nearest excavations were Dr Rowe's work in the Half Mile Ride
Anglo-Saxon cemetery (1923) which is 700 m. south of the site, the
Thanet Trust's evaluations at Westfield Road about 300 m. to the west,
and Dr Rowe's 'Tivoli Villa' about 300 m. north-east. A crop-mark soilmark
that may well be significant has been recorded from the southwestern
rise of the Shottendane valley as it ascends towards St. John's
cemetery. It seems to indicate the presence of double parallel ditches
passing along the face of the hill and turning east towards the Lydden
valley. They may constitute the defences of a prehistoric hill-fort similar
to the great enclosure recently discovered at North Foreland Hill,
Kings gate.
Historical Background
The origin of the name Hartsdown is as a hill frequented by harts or
stags (OE herotes dun) and the hill was known as 'Hertesdowne' in 1451.
This apart, the tract of land is seldom mentioned during the medieval
period, when it would have formed part of the Manor of Dent-de-Lion.
From the 1750s the land was in use as arable and pasture, and was the
site of the 'Margate Races' from 1765 to 1838.
METHODS
Trenching
Evaluation was by machine-cut trench. In all, 59 trenches were cut under
archaeological supervision, in total 2450 m. of trenching with an area of
3675 square metres, this representing a 2 per cent sampling of the site.
As originally planned, twenty-three of the trenches were so positioned
as to coincide with crop-marks, whose location had been established by
photogrammetry during preliminary work by the Thanet Trust in 1991.
The rest of the trenches was positioned so as to sample 'bare' areas
where no crop-marks had been observed so as to act as a control. For
267
D. R. J. PERKINS
safety reasons, trenches were backfilled at the end of each working session.
Important features which could not be excavated or recorded in a
day were left covered and taped off. All trenches were cut under archaeological
supervision, and were taken down either to an ancient horizon or
the natural chalk. Any isolated features were excavated, and where the
trenches ran through settlement enclosures about 25 per cent of the
features observed were, according to size, either sampled or fully
excavated.
It had been intended that rather more than double the amount of
trenching would be carried out so as to sample at about the 4 per cent
level. In the event, although an additional 650 square metres of unscheduled
trenches were cut so as to exactly locate sites and define the boundaries
of the settlement enclosures, the final total (about 2.4 per cent) is
thought adequate, see below.
Confidence Rating
This can be considered high. In the case of those trenches whose location
coincided with the crop-mark evidence, the features were found. In all
others, the 'random trenches' where no crop-marks were visible in the
air photos, nothing was observed. Additionally, the shallow overburden
covering most of the site promotes the formation of crop-marks, even
over small and shallow features such as post-holes. The assembled cropmark
picture can therefore be considered explicit.
RESULTS: SITE DESCRIPTIONS
The following is a list of 'Sites' where a group of features is either
confined within a boundary ditch, or are in such close association as to
be considered a single archaeological entity. Fig. 1 shows the location
of all trenches and sites. Ceramic spot dates are as given by Nigel
Macpherson-Grant, who has examined the Hartsdown pottery and will
be commenting on it in due course as part of a synthesis of Late Bronze
Age-Early Iron Age ceramics in east Kent.
Site 1
This is clearly identifiable as a length of Roman metalled trackway. It
consists of a layer of metalling composed of small water-worn flint cobbles
about 4 m. wide and 0.20 m. deep where best preserved, this laid
in a shallow chalk-cut slot. This is flanked on both sides by parallel
chalk-cut ditches about 4 m. wide and 1 m. deep. In the only complete
section cut (Trench 21) these have been re-cut at some stage so as to
create double ditches on either side (see Fig. 2, plan and section). The
268
'°
Site 13
Site 12 4 Site 7 Site 5 ·. '✓.
"
ite 3 v
\ L(' -., \'' %
--------=- -----:---- --- 1--
-( ' --- :--- I
--.\ -
--- I {I-}-, --- I
I • 8 -T:. ' 1 I - - --- *: Site -1 , __ _
I '
-- - ---
_ _ ------- Site 6
-"'-1/. \ 1'o.
1
- , Site J 'i>'c,,
___ ; -- {),. \ \ -·-\- \ .
I S\\.e,
\ ,)- μ
1/
Site2
.- -:- --· i '
--- , Site I • -- _ S,te 9
•:----+-. -
------------ l
l --- i j ___ C-- ' • • I --
--t=il-
Site IO r- ! ' ta ww■■ 200m.
' '
R Idge.r
doWnl. l!Je or
aJldeS carp l!Jem
Fig. 1. Hartsdown Evaluation 1995, Plan of Trenches and Sites. Trenches are marked in broken line, and crop marks are shown
in heavy line for the Roman trackway and sites not illustrated in Figs. 2-5.
::r:
t:I
(j
0
I
0
0
;
V, (j
j
a)
b)
\
\
\
/
/
'
\
/
/
/
/
/
\
'
'
✓
/
✓
/
/
' -----,-.=R=•-
Preserved ancient horizon
:2
N
Trench 21a)
r---c.-= l <-:=-.:
I; I (
',
I,
,', , ,
, ,
,
,
✓
,
/
,
I
'
I
Trench 21d) !Orn. -----